The Paper 1 has reported the original X-ray magnetic circular dichroism as follows: a magnetic sample is placed in the external strong magnetic field, an internal magnetization direction of the magnetic sample is aligned with one direction, then the sample is irradiated with circular polarization X-ray while the external magnetic field direction is alternately inverted, resultantly, intensity of transmission X-ray passing the sample is changed in accordance with whether the external magnetic field direction is parallel or anti-parallel to the polarization light direction, the intensity change appears remarkably at a neighborhood of X-ray absorption edge for the constituent element of the magnetic sample. Applying the above X-ray magnetic circular dichroism, very old method such as the patent literature 1 has been presented to observe magnetic samples. The patent literature 1 presented the method and apparatus to observe magnetic domain by X-ray magnetic circular dichroism (XMCD) as follows: the intensity (I0) of incident X-ray and intensity (It) of transmission X-ray passing the sample are measured, μR×t=ln(I0/It) and μL×t =ln(I0/It) are calculated using the observed intensity (I0) and observed intensity (It), M=(μR×t−μL×t)/(μR×t+μL×t)=(μR−μL)/(μR+μL) is calculated using [μR×t] and [μL×t], and M is transformed into electronic signal for imaging, here, μR and μL are X-ray absorption coefficient for right-circular polarization X-ray and left-circular polarization X-ray, respectively, and t is thickness of the sample. However the method was impracticable because focusing of X-ray is no good, sensitivity of X-ray detection is very low, and X-ray transmittance for right-circular polarization X-ray and left-circular polarization X-ray are calculated from observed X-ray intensity ln(I0/It) but indirectly observed.
Observation, structural analysis and elemental analysis of the structures and chemical and physical states of magnetic sample surfaces or their neighborhood have been studied using the secondary electrons which are emitted from the sample surfaces by irradiating the observed areas of the sample with electron beams or exciting light. For example, the patent literature 2 presented the method to form observation images of microscopic structures of the sample. However, the above conventional method has included several difficult problems: (1) the conventional method is merely able to observe magnetic characteristics in the region from uppermost surfaces of the sample to several nanometers in depth but unable to observe them over several nanometers in depth because any electrons to generate from places deeper than several nanometers cannot get out of the sample surfaces. Because, the method is the one to detect the secondary electrons that are emitted from the magnetic sample on absorption of synchrotron radiation light, (2) the conventional method has essentially undesirable problem that the observed analytical result does not always show true magnetic properties of the bulk due to interruption effect of oxidized layers in the uppermost surfaces, (3) the conventional method is undesirable to observe the magnetic sample within magnetic field because of an applied magnetic field to exert influence on the detection of the secondary electrons. From these reasons, it has been very difficult to measure in precise the magnetized structure in the inside of micro-particles.